|
Assessment of nutrient intakes of picky-eating Chinese preschoolers using a modified food frequency questionnaire
Flora Yin-Ying Kwok, Yvonne Yi-Fong Ho, Chung-Mo Chow, Christabella Yat-Nam So, Ting-Fan Leung
Hong Kong, China
Author Affiliations: Department of Pediatrics, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong (Kwok FY, Ho YY, Chow CM, Leung TF); University of Western Ontario, London, Ontario, Canada (So CY)
Corresponding Author: Ting Fan Leung, MD, Department of Pediatrics, 6/F, Clinical Sciences Building, Prince of Wales Hospital, Shatin, N.T., Hong Kong (Tel: 852-2632 2981; Fax: 852-2636 0020; Email: tfleung@cuhk.edu.hk))
doi: 10.1007/s12519-012-0386-9
Background: Dietary assessment is crucial for monitoring nutritional status of young children. This study applied a modified Chinese food frequency questionnaire (FFQ) to assess nutrient intakes of young picky-eating Hong Kong children.
Methods: Nutrient intakes were obtained by FFQ and 3-day food record (3DFR) in 29 picky-eating children aged 44.8㊣9.2 months who participated in a randomised controlled trial of a new milk formula.
Results: When compared with 3DFR, FFQ over-reported energy intake by 283 kcal (26.5%) at baseline and 237 kcal (21.4%) at end-of-study, and also over-estimated intakes of carbohydrates, proteins and fats. At baseline, FFQ and 3DFR classified 34.4%-65.5% of subjects into the same tertiles for most nutrients. These methods showed weak-to-moderate agreement in measuring most nutrients, with 3DFR showing a trend towards a systematic increase in the differences with increasing nutrient intake.
Conclusion: Our FFQ shows promising results for assessing nutrient intakes in picky-eating Chinese preschoolers.
Key words: Chinese; food frequency questionnaire; nutrition; preschooler
World J Pediatr 2013;9(1):58-63
Introduction
Dietary assessment is crucial for monitoring the nutritional status of children, examining association between diet and health, and identifying dietary factors that are associated with individual clinical conditions. However, measuring diet in very young children is difficult because of the rapid changes in food habits of toddlers, the need to rely on parental report, and the questionable ability of parents to accurately report their children's diet when other caregivers, such as at day care, also feed the child. Food frequency questionnaire (FFQ) reported by parents appears to be a reasonably valid method to collect dietary data in children even in situations when parents do not observe all meals and snacks eaten by their children.[1,2] Most published FFQs were designed and validated in Caucasian populations.[2-8] Population-specific FFQ must be used due to wide variations in dietary habits between different populations. Woo et al[9] developed an FFQ for Chinese adults in Hong Kong, which has subsequently been adapted for use in school-age children and elderly.[10-12] This study adapted the above Chinese FFQ appropriately to reflect the portion sizes typically consumed by young children and applied the modified FFQ to assess nutrient intakes in picky-eating Hong Kong preschool children. A 3-day food record (3DFR) was used as our reference method.
Methods
Subjects and study design
Subjects were Chinese children aged 2-5 years from local kindergartens who participated in a multi-center, randomized controlled trial that studied the efficacy of 4-month treatment with a balanced cow's milk formula.[13] The subjects were reported by parents to have picky-eating behaviour, and with weight-for-height indicator ≒25th percentile according to the World Health Organisation Child Growth Standards (http://www.who.int/childgrowth/en/). Subjects' parents completed FFQ and 3DFR at baseline and 4 months later.
Our FFQ consisted of 262 items categorised into 10 groups: bread/cereals/pasta/rice/noodles, vegetables and legumes, fruits, meat, fish, eggs, milk and dairy products, beverages, dim sum/snacks/fats/oils, and soups. Parents completed this FFQ with the assistance from trained research staff. The options for frequency varied from "never/less than once per month" to "several times a day". Three dimensional food models were provided for parents to aid their estimation of the portion sizes. Our modified FFQ was pilot-tested for readability in five preschool children with allergies before use in this study.
At each time point, the 3DFR covered two weekdays and one weekend. All subjects were instructed to follow their usual diet on those 3 days. Parents were provided with record notebooks and photographic booklets that included colour photographs of common food items and of household measuring units such as differently sized bowls, plates, spoons, cups and glasses to aid the estimation of food portions. Parents were also asked to provide detailed description of each food, including brand name or method of preparation and recipes, whenever possible.
The recording of food intake by 3DFR started 5-7 days after FFQ at baseline, and the sequence was reversed with 3DFR completed 5-7 days ahead of FFQ at 4 months. The Clinical Research Ethics Committee of our University approved this study, and informed written consent was obtained from the parents of the studied children.
Dietary assessment methods
Our 262-item FFQ incorporated appropriate changes of the initial FFQ[9] to reflect portion sizes typically consumed by young children. Because dietary intake is highly variable at this age, parents were asked to think of their children's diet over the past one month when they filled in such questionnaire. 3DFR based on estimated food intakes covered two weekdays and one weekend. All subjects were instructed to follow their usual diet on those 3 days.
Nutrient calculations
Foods and nutrient intakes obtained from FFQ and 3DFR were computed by the Foodworks nutrient analysis software (Professional version 6; Xyris Software, Highgate Hill, Queensland, Australia), with the addition of some local foods based on food composition tables from China.[14,15]
Statistical analysis
Numerical variables were compared between the two dietary assessment methods by Wilcoxon's signed-rank test, whereas the correlations between the two measurements were analysed by Spearman's rank-order correlation coefficients. The agreement between the two dietary assessment methods was analysed by Bland-Altman plots, and the agreement between categorical parameters was examined by kappa coefficients following the classification of absolute nutrient intakes into tertiles. All analyses were performed by SPSS v.17 (SPSS Inc., Chicago, IL, USA), and P values <0.05 were considered to be statistically significant.
Results
Intakes of individual nutrients by FFQ and 3DFR
Twenty-nine subjects (Table 1) returned both FFQ and 3DFR at baseline, and two boys and two girls were lost to follow-up. Table 2 summarizes the daily absolute intakes of energy and nutrients based on FFQ and 3DFR at baseline. FFQ over-reported the energy intake in our subjects by 283 kcal (26.5%). Generally, medium to high correlations were found in most nutrients (Spearman's rank-order correlation coefficient, 老>0.29). Table 3 summarizes the daily absolute intakes of energy and nutrients based on FFQ and 3DFR at study completion. FFQ over-reported energy intake by about 237 kcal (21.4%). This method also yielded higher estimates for intakes of many other nutrients. No significant differences were found between the estimates of intakes of other nutrients by these two methods. Medium to high correlations were found between FFQ and 3DFR in all nutrients (老>0.35) except riboflavin (老=0.255).
Agreement for nutrient intakes by FFQ and 3DFR
Table 4 summarizes the percentages of subjects who were correctly classified and misclassified for nutrient intakes at baseline and study completion. At the baseline, the two methods classified 34.4%-65.5% of subjects into the same tertiles for most nutrients. Gross misclassification (from one extreme tertile to the other extreme) occurred at a median of 17.2% (range: 6.8%-20.7%). FFQ and 3DFR showed weak to moderate agreement (百肋) in the measurements of most nutrients (ranged from 百肋=0.016 for energy intake from carbohydrate to 百肋=0.482 for iron). At the study completion, FFQ and 3DFR classified 36%-64% of subjects into the same tertile. Gross misclassification occurred at a median of 8.0% (range: 0-12.0%). The two methods again showed weak to moderate agreement for all nutrients (ranged from 百肋=0.038 for carbohydrate and dietary fibres to 百肋=0.459 for zinc). The agreement for iron intakes was satisfactory at the baseline and the end of the study.
Figs. 1 and 2 are Bland-Altman plots for the agreement in selected nutrients by FFQ and 3DFR. There was a trend towards a systematic increase in the differences between FFQ and 3DFR with increasing intake. Our observed differences suggested that participants both under-reported and over-reported nutrient intakes with FFQ compared with 3DFR.
Discussion
Vereecken and colleagues[8] compared food intakes by a 77-item FFQ with an online 3DFR in 216 Belgian-Flemish preschool children. At the group level, they identified good agreement for energy, fat and protein intake but an overestimation of carbohydrates (5.6%) and dietary fibre (13.3%) and an underestimation of calcium (9%). Our FFQ significantly overestimated energy, carbohydrate, fat and dietary fibre compared with 3DFR at both the baseline and the end of 4-month follow-up (Tables 1 and 2). Such overestimation may be explained by recall bias of parents and their choice of food items. If a particular food contains more than one item under the FFQ food list, it is possible for parents to report those items and overestimate the relevant nutrient intakes.
Calcium intake is crucial for the long-term bone health of young children. Taylor et al[16] compared calcium intake from a 35-item short calcium FFQ with that from a 4-day diet record. They concluded that the short calcium FFQ would not be appropriate for determining calcium intake of young children because it overestimated actual calcium intakes. A subsequent study found that FFQ underestimated preschool children's calcium intake (838 by 3DFR and 777 by FFQ).[17] More than 50% of our picky-eating subjects took lower calcium than the US age-specific RDA (800 mg/day; Table 1). In addition, although we found moderate correlation for calcium intakes between FFQ and 3DFR (Spearman's rank-order correlation coefficient, 老>0.537), our FFQ overestimated calcium intake when compared with 3DFR at the baseline (median, 609.6 mg/day vs. 459.3 mg/day) and the end of the study (median, 760.8 mg/day vs. 579.4 mg/day).
Rogers et al[18] investigated the long-term implications of nutritional assessments throughout childhood in over 3000 British girls. Diet was assessed by FFQ at 3 and 7 years of age, and by 3DFR at 10 years. They found intakes of total and animal protein as well as polyunsaturated fatty acids at 3 and 7 years to be positively associated with earlier age at menarche. Such results may alter the lifetime risks of breast cancer and osteoporosis, thus supporting the importance of using either FFQ or food diary to assess nutrient intakes in preschool children.
This study has several limitations. Bland-Altman plots revealed wide limits of agreement of the differences in nutrient intakes between FFQ and 3DFR. This observation could be explained by the small number of subjects being recruited. Our results need to be confirmed in larger studies. Besides, the subjects had subnormal body weight (BMI z-scores -0.57 to -3.12) and abnormal eating behaviours, so this study could not reflect dietary intakes in healthy children. Our participants had picky-eating behaviour. A good agreement of the tools in our subjects might not be generalisable to the general preschool population. In addition, our FFQ should ideally be validated against doubly-labelled water.[19,20] Because of time and financial constraints, however, we consider it acceptable to compare FFQ with food record as in many published studies.[2-8,18,20]
In conclusion, our modified FFQ is an acceptable method for assessing intakes of energy, macronutrients and selected vitamins and minerals in picky-eating Chinese preschoolers. This method is more useful in assigning overall ranking than assessing the actual nutrient intakes. Our findings also need to be confirmed in larger community-based cohorts.
Acknowledgement
We thank S. S. Wang and H. Y. Sy for their help with data collection and entry.
Funding: This study was funded by Direct Grant for Research of the Chinese University of Hong Kong.
Ethical approval: This study was approved by Clinical Research Ethics Committee of The Chinese University of Hong Kong (reference CRE-2009.637-T).
Competing interest: The randomized controlled clinical trial for the balanced milk formula was sponsored by Shanghai Wyeth Nutritional under protocol number 9061A1-4000. However, the sponsor was not involved in the design of this study, collection and analysis of data and the preparation of this manuscript.
Contributors: Kwok FY and Ho YY collected anthropometric and nutritional data and helped in data analyses; Chow CM participated in the study design and collected clinical data; So CY administered food questionnaires; Leung TF designed the study, collected, analyzed and interpreted data and wrote the manuscript.
References
1 Thompson FE, Byers T. Dietary assessment resource manual. J Nutr 1994;124(11 Suppl):2245S-2317S.
2 Andersen LF, Lande B, Trygg K, Hay G. Validation of a semi-quantitative food-frequency questionnaire used among 2-year-old Norwegian children. Public Health Nutr 2004;7:757-764.
3 Wilson AM, Lewis RD. Disagreement of energy and macronutrient intakes estimated from a food frequency questionnaire and 3-day diet record in girls 4 to 9 years of age. J Am Diet Assoc 2004;104:373-378.
4 Klohe DM, Clarke KK, George GC, Milani TJ, Hanss-Nuss H, Freeland-Graves J. Relative validity and reliability of a food frequency questionnaire for a triethnicpopulation of 1-year-old to 3-year-old children from low-income families. J Am Diet Assoc 2005;105:727-734.
5 Huybrechts I, De Backer G, De Bacquer D, Maes L, De Henauw S. Relative validity and reproducibility of a food-frequency questionnaire for estimating food intakes among Flemish preschoolers. Int J Environ Res Public Health 2009;6:382-399.
6 Ovaskainen ML, Nevalainen J, Uusitalo L, Tuokkola JJ, Arkkola T, Kronberg-Kippilä C, et al. Some similarities in dietary clusters of pre-school children and their mothers. Br J Nutr 2009;102:443-452.
7 Huybrechts I, Vereecken C, De Bacquer D, Vandevijvere S, Van Oyen H, Maes L, et al. Reproducibility and validity of a diet quality index for children assessed using a FFQ. Br J Nutr 2010;104:135-144.
8 Vereecken C, Covents M, Maes L. Comparison of a food frequency questionnaire with an online dietary assessment tool for assessing preschool children's dietary intake. J Hum Nutr Diet 2010;23:502-510.
9 Woo J, Leung SS, Ho SC, Lam TH, Janus ED. A food frequency questionnaire for use in the Chinese population in Hong Kong: description and examination of validity. Nutr Res 1997;17:1633-1641.
10 Sea MM, Woo J, Tong PC, Chow CC, Chan JC. Associations between food variety and body fatness in Hong Kong Chinese adults. J Am Coll Nutr 2004;23:404-413.
11 Yu CC, Sung RY, Hau KT, Lam PK, Nelson EA, So RC. The effect of diet and strength training on obese children's physical self-concept. J Sports Med Phys Fitness 2008;48:76-82.
12 Chan RS, Woo J, Chan DC, Cheung CS, Lo DH. Estimated net endogenous acid production and intake of bone health-related nutrients in Hong Kong Chinese adolescents. Eur J Clin Nutr 2009;63:505-512.
13 Yao M, Sheng X, Tong M, Zhao D, Leung TF, Zhang F, et al. Effects of nutrition counseling with and without growing-up milk on growth and micronutrient adequacy in picky eaters: a randomized controlled trial. Evid-Based Child Health 2011;6(Suppl 1):102.
14 Yang Y, Wang G, Pan X. China Food Composition Table 2002. Beijing: Peking University Medical Press, 2002.
15 Yang Y. China Food Composition Table 2004 (Book 2). Beijing: Peking University Medical Press, 2005.
16 Taylor RW, Goulding A. Validation of a short food frequency questionnaire to assess calcium intake inchildren aged 3 to 6 years. Eur J Clin Nutr 1998;52:464-465.
17 Huybrechts I, De Bacquer D, Matthys C, De Backer G, De Henauw S. Validity and reproducibility of a semi-quantitative food-frequency questionnairefor estimating calcium intake in Belgian preschool children. Br J Nutr 2006;95:802-816.
18 Rogers IS, Northstone K, Dunger DB, Cooper AR, Ness AR, Emmett PM. Diet throughout childhood and age at menarche in a contemporary cohort of British girls. Public Health Nutr 2010;13:2052-2063.
19 Davies PS, Coward WA, Gregory J, White A, Mills A. Total energy expenditure and energy intake in the pre-school child: a comparison. Br J Nutr 1994;72:13-20.
20 McPherson RS, Hoelscher DM, Alexander M, Scanlon KS, Serdula MK. Dietary assessment methods among school-aged children: validity and reliability. Prev Med 2000;31:S11-S33.
Received August 13, 2011 Accepted after revision November 8, 2011
|
